[R-390] JAN vs JHS spec tubes, mixer specs, etc.

[Tim Shoppa]

After buying some 6C4WA's recently, of course I found a huge
stash (technically measured as a "metric buttload") of 6C4W's
and 6C4WA's in my crawlspace. I probably bought them years
ago and forgot about them.

That's the nice thing about chronic forgetfulness - every time I
look somewhere I don't see everyday, I'm pleasantly surprised
by stuff that I had forgotten it even existed. It's like Christmas
year-round, and every day I get a new train set!

Anyway, this stash has the traditional mil-spec white boxes,
but the tubes (Sylvanias) are labeled twice (once in grey, once
in green) and the boxes declare them to be "JHS" instead
of "JAN".

Can someone remind me what JHS means? If I asked this a
few months ago when I last saw these boxes, you can remind
me again, it'll be like I never heard the answer first time around!

And some further thoughts about how these tubes are used:

I recently opened up my copy of Seeley's "Electron Tube Circuits"
and was looking through the section on mixers. There he makes
the comment that anything that lowers the gain of a mixer can
greatly impact conversion efficiency, and this is related to the
DC operating point and AC injection levels of the mixer.

This got me thinking of the three mixers in the 390A:

1. Each has a 27 ohm resistor on the grid, where the
signal is fed in. The DC level on the grid comes from the
AGC line (except for the 3rd mixer).
2. Each has a 2.2K resistor on the plate and a tuned
circuit for extracting the output.
3. Each has a (bypassed) 2.2K resistor on the cathode, through
which the oscillator feeds in. The oscillator outputs are all
transformer coupled so the bottom of the 2.2K is at DC
ground.

This exact circuit is replicated 3 times. Looking at the schematics
the repetition is almost monotonous. Anyone care to
comment on the circuit? The 3-time repitition looks like some
kind of holy mantra this morning. My experience probing around with
a scope is that the cathodes get several volts of
oscillator signal. If I wanted to translate this circuit and
levels to a conversion efficiency, I'd start by finding the
voltages and currents for various grid biases (AGC voltages),
and then reading the transconductance off the curves, right?

Where does the 27 ohm resistor on the grid come in? The DC
impedances around the grid are in the Megohms, I don't really
understand (off the top of my head) how the 27 ohm resistor
does much of anything. Unless it is there to add a little bit
of AC impedance so the previous stage's plate circuit doesn't
get detuned by grid current? In which case the AC impedances
must be in the tens of ohms? Isn't this a bit on the low side for
a typical RF grid or plate circuit?

Tim.